The most characteristic change in progressive dementia of Alzheimer's type is a tissue deposit of amyloid beta/A4 protein, which is derived from its precursor protein APP (ref.2). Structural alterations of APP are implicated in the pathogenesis of Alzheimer's disease, but it is not known how they cause the disease. Although APP has a receptor-like architecture, is located on the neuronal surface, and has a conserved cytoplasmic domain, no receptor function has been demonstrated for APP. Here we report that APP forms a complex with G(o), a major GTP-binding protein in brain. The cytoplasmic APP sequence His 657-Lys 676 shows a specific G(o)-activating function and is necessary for complex formation. G(o) protein treated with GTP-gamma S lost the ability to associate with APP. This suggests that APP is a receptor coupled to G(o) and that abnormal APP-G(o) signalling is involved in the Alzheimer's disease process.

Amyloid precursor protein (APP), a transmembrane precursor of beta-amyloid, possesses a function whereby it associates with G(o) through its cytoplasmic His657-Lys676. Here we demonstrate that APP has a receptor function. In phospholipid vesicles consisting of baculovirally made APP695 and brain trimeric G(o), 22C11, a monoclonal antibody against the extracellular domain of APP, increased GTP gamma S binding and the turnover number of GTPase of G(o) without affecting its intrinsic GTPase activity. This effect of 22C11 was specific among various antibodies and was observed neither in G(o) vesicles nor in APP695/Gi2 vesicles. In APP695/G(o) vesicles, synthetic APP66-81, the epitope of 22C11, competitively antagonized the action of 22C11. Monoclonal antibody against APP657-676, the G(o) binding domain of APP695, specifically blocked 22C11-dependent activation of G(o). Therefore, APP has a potential receptor function whereby it specifically activates G(o) in a ligand-dependent and ligand-specific manner.

◥ Accumulating evidence indicates that CD8 þ T cells in the tumor microenvironment and systemic CD4 þ T-cell immunity play an important role in mediating durable antitumor responses. We longitudinally examined T-cell immunity in the peripheral blood of patients with non-small lung cancer and found that responders had significantly (P < 0.0001) higher percentages of effector, CD62L low CD4 þ T cells prior to PD-1 blockade. Conversely, the percentage of CD25 þ FOXP3 þ CD4 þ T cells was significantly (P ¼ 0.034) higher in nonresponders. We developed a formula, which demonstrated 85.7% sensitivity and 100% specificity, based on the percentages of CD62L low CD4 þ T cells and CD25 þ FOXP3 þ cells to predict nonresponders. Mass cytometry analysis revealed that the CD62L low CD4 þ T-cell subset expressed T-bet þ , CD27 À , FOXP3 À , and CXCR3 þ , indicative of a Th1 subpopulation. CD62L low CD4 þ T cells significantly correlated with effector CD8 þ T cells (P ¼ 0.0091) and with PD-1 expression on effector CD8 þ T cells (P ¼ 0.0015). Gene expression analysis revealed that CCL19, CLEC-2A, IFNA, IL7, TGFBR3, CXCR3, and HDAC9 were preferentially expressed in CD62L low CD4 þ T cells derived from responders. Notably, longterm responders, who had >500-day progression-free survival, showed significantly higher numbers of CD62L low CD4 þ T cells prior to PD-1 blockade therapy. Decreased CD62L low CD4 þ T-cell percentages after therapy resulted in acquired resistance, with longterm survivors maintaining high CD62L low CD4 þ T-cell percentages. These results pave the way for new treatment strategies for patients by monitoring CD4 þ T-cell immune statuses in their peripheral blood.

No abstract